TaPERA: Enhancing Faithfulness and Interpretability in Long-Form Table QA by Content Planning and Execution-based Reasoning
Yilun Zhao, Lyuhao Chen, Arman Cohan, Chen Zhao
摘要
Long-form Table Question Answering (LFTQA) requires systems to generate paragraph long and complex answers to questions over tabular data. While Large language models based systems have made significant progress, it often hallucinates, especially when the task involves complex reasoning over tables. To tackle this issue, we propose a new LLM-based framework, TAPERA, for LFTQA tasks. Our framework uses a modular approach that decomposes the whole process into three sub-modules: 1) QA-based Content Planner that iteratively decomposes the input question into sub-questions; 2) Execution-based Table Reasoner that produces executable Python program for each sub-question; and 3) Answer Generator that generates long-form answer grounded on the program output. Human evaluation results on the FETAQA and QTSUMM datasets indicate that our framework significantly improves strong baselines on both accuracy and truthfulness, as our modular framework is better at table reasoning, and the long-form answer is always consistent with the program output. Our modular design further provides transparency as users are able to interact with our framework by manually changing the content plans. https://github.com/yilunzhao/TaPERA Plan-based Answer Generation Direct Answer Generation Q1: Which company earns the highest profit in the Oil and Gas industry? A1: Sinopec Group earns the highest profit in the Oil and Gas industry. Q2: Which company earns the overall highest profit? A2: Apple earns the overall highest profit. Q3: Compare these two companies. A3: [pending] Q1: Which company earns the highest profit in the Oil and Gas industry? A1: Sinopec Group earns the highest profit in the Oil and Gas industry. Q2: Which company earns the overall highest profit? A2: Apple earns the overall highest profit. Q3: Compare these two companies. A3: Apple is in the electronics industry, while Sinopec Group is in the Oil and Gas industry.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper11
- Table-Critic: A Multi-Agent Framework for Collaborative Criticism and Refinement in Table ReasoningPeiying Yu, Guoxin Chen, Jingjing WangACL 2025 · 被引用 30 次
- When TableQA Meets Noise: A Dual Denoising Framework for Complex Questions and Large-scale TablesShenghao Ye, Yu Guo, Dong Jin, Yuxiang Wang 等ACL 2026 · 被引用 8 次
- MT-RAIG: Novel Benchmark and Evaluation Framework for Retrieval-Augmented Insight Generation over Multiple TablesKwangwook Seo, Donguk Kwon, Dongha LeeACL 2025 · 被引用 8 次
- Reinforced Informativeness Optimization for Long-Form Retrieval-Augmented GenerationYuhao Wang, Ruiyang Ren, Yucheng Wang, Xin Zhao 等ACL 2026 · 被引用 4 次
- Table Question Answering in the Era of Large Language Models: A Comprehensive Survey of Tasks, Methods, and EvaluationWei Zhou, Bolei Ma, Annemarie Friedrich, Mohsen MesgarACL 2026 · 被引用 3 次
它引用的顶会 Paper25
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma 等NeurIPS 2022 · 被引用 22,562 次
- BERTScore: Evaluating Text Generation with BERTTianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger 等ICLR 2020 · 被引用 8,443 次
- Toolformer: Language Models Can Teach Themselves to Use ToolsTimo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu 等NeurIPS 2023 · 被引用 5,989 次
- TabFact: A Large-scale Dataset for Table-based Fact VerificationWenhu Chen, Hongmin Wang, Jianshu Chen, Yunkai Zhang 等ICLR 2020 · 被引用 674 次
- TAPEX: Table Pre-training via Learning a Neural SQL ExecutorQian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi 等ICLR 2022 · 被引用 347 次
相关 Paper
- Generation-Focused Table-Based Intermediate Pre-training for Free-Form Question AnsweringPeng Shi, Patrick Ng, Feng Nan, Henghui Zhu 等AAAI 2022 · 被引用 16 次
- Chain-of-Table: Evolving Tables in the Reasoning Chain for Table UnderstandingZilong Wang, Hao Zhang, Chun-Liang Li, Julian Martin Eisenschlos 等ICLR 2024 · 被引用 244 次
- Accurate and Regret-Aware Numerical Problem Solver for Tabular Question AnsweringYuxiang Wang, Jianzhong Qi, Junhao GanAAAI 2025 · 被引用 12 次
- AixelAsk: A Stepwise-Guided Retrieval and Reasoning Framework for Large Table QAChi Zhang, Meihui Zhang, Yuxin Yang, Tao Chen 等SIGMOD 2026 · 被引用 2 次
- Large Language Models are Versatile Decomposers: Decomposing Evidence and Questions for Table-based ReasoningYunhu Ye, Binyuan Hui, Min Yang, Binhua Li 等SIGIR 2023 · 被引用 75 次
